Hello Jack, clear thanks. But are you able to reproduce the issue sistematically? If yes, in theory, if you restart the slower broker instance and you let it get the lease again (e.g. shutting down all the others), it should flush in the shared datastore the still pending messages it has still got inside its local datastore. Anyway it's hard to say if such error is related to the local or to the shared datastore.
*Distinti Saluti / *Kind Regards M.G. Il giorno lun 26 feb 2024 alle ore 11:23 Jack Fry (They/Them) <j...@scottlogic.com.invalid> ha scritto: > Hi Marco, > > Thanks for getting back to me, we're really scratching our heads over this > one. The second log frame is on a different broker to the first log frame. > We assume the first broker is creating the record but not sending the > message, then the second broker cannot create the record and cannot send > the message. > > Related to the local data store, we've noticed an error: > > [glencore_activemq.log][WARN] Store COMMIT FAILED: > [glencore_application_logging.log][Error] javax.transaction.xa.XAException > : STORE COMMIT FAILED: Transaction rolled back > > Is this coming from the local datastore? > > Many thanks, > Jack >
